Adaptive Reasoning Executor: A Collaborative Agent System for Efficient Reasoning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908594078744576 |
|---|---|
| author | Ling, Zehui Chen, Deshu Zhang, Yichi Liu, Yuchen Li, Xigui Guo, Xin Cheng, Yuan |
| author_facet | Ling, Zehui Chen, Deshu Zhang, Yichi Liu, Yuchen Li, Xigui Guo, Xin Cheng, Yuan |
| contents | Recent advances in Large Language Models (LLMs) demonstrate that chain-of-thought prompting and deep reasoning substantially enhance performance on complex tasks, and multi-agent systems can further improve accuracy by enabling model debates. However, applying deep reasoning to all problems is computationally expensive. To mitigate these costs, we propose a complementary agent system integrating small and large LLMs. The small LLM first generates an initial answer, which is then verified by the large LLM. If correct, the answer is adopted directly; otherwise, the large LLM performs in-depth reasoning. Experimental results show that, for simple problems, our approach reduces the computational cost of the large LLM by more than 50% with negligible accuracy loss, while consistently maintaining robust performance on complex tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_13214 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adaptive Reasoning Executor: A Collaborative Agent System for Efficient Reasoning Ling, Zehui Chen, Deshu Zhang, Yichi Liu, Yuchen Li, Xigui Guo, Xin Cheng, Yuan Artificial Intelligence Recent advances in Large Language Models (LLMs) demonstrate that chain-of-thought prompting and deep reasoning substantially enhance performance on complex tasks, and multi-agent systems can further improve accuracy by enabling model debates. However, applying deep reasoning to all problems is computationally expensive. To mitigate these costs, we propose a complementary agent system integrating small and large LLMs. The small LLM first generates an initial answer, which is then verified by the large LLM. If correct, the answer is adopted directly; otherwise, the large LLM performs in-depth reasoning. Experimental results show that, for simple problems, our approach reduces the computational cost of the large LLM by more than 50% with negligible accuracy loss, while consistently maintaining robust performance on complex tasks. |
| title | Adaptive Reasoning Executor: A Collaborative Agent System for Efficient Reasoning |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2510.13214 |